Despite the control of HIV replication and the decreased incidence of AIDS due to HAART, large cohort studies report that mild/moderate cognitive impairment occurs in up to 50 % of those on long term HAART. Further, it is not clear whether eariy versus delayed initiation of HAART impacts the CNS or the development of cognitive impairment. The CNS is a reservoir for HIV latency and it is unclear how reactivation in the CNS contributes to residual viremia in the periphery or the ongoing evolution of the virus in brain and in the periphery of HAART treated individuals. Intensification of current HAART regimens does not appear to reduce the latent reservoir in resting CD4+ lymphocytes and is unlikely to diminish tissue reservoirs of HIV, particulariy the CNS. Thus, strategies for reducing and purging latent reservoirs are needed because of the high cost and the known (peripheral neuropathy and other side effects) and unknown effects of lifelong therapy. Structured treatment interruption has shown that these approaches increased the latent reservoir or expanded the quasi-species of virus in an individual, cleariy not benefiting the patient and in some causing activation of HIV in the CNS. Thus, novel approaches for reducing or purging the latent reservoir have the potential for adverse effects on HAART. Testing of therapeutic approaches would be best done in an animal model that recapitulates latent viral reservoirs in all tissues infected in HIV. Our SIV HAART model provides a rigorous model to examine many aspects of HIV latency and to test novel therapy to reduce or purge latent reservoirs in CNS and other tissues. Project 2 will examine whether when HAART is initiated (acute infection, post-acute infection of eariy disease) impacts infection and latency in brain by quantitating latently infected cells in the brain. Another measure of viral latency is the presence of residual virus replication and virus evolution, residual virus in the CSF will be studied longitudinally as a measure of ongoing virus replicafion in brain when virus is suppressed in the peripheral blood. Finally, viral latency in the brain has not been well characterized and it is important to know the molecular state of the viral DNA in order to select appropriate drug candidates for activating latent virus in tissue for strategies that reduce or purge the latently infected cells without virus spread. We will examine histone modifications in latent virus in vivo to determine whether HDAC inhibitors or other drugs will be effective in altering the state of virus DNA from quiescent to active in brain and spleen. This project provides comprehensive analyses of the viral DNA in the brain with regard to molecular state, level of reactivation and evolution and the epigenetic modifications that exist during infection and treatment with HAART.